Putting the 'role' back in role-playing games since 2002.
Donate to Codex
Good Old Games
  • Welcome to rpgcodex.net, a site dedicated to discussing computer based role-playing games in a free and open fashion. We're less strict than other forums, but please refer to the rules.

    "This message is awaiting moderator approval": All new users must pass through our moderation queue before they will be able to post normally. Until your account has "passed" your posts will only be visible to yourself (and moderators) until they are approved. Give us a week to get around to approving / deleting / ignoring your mundane opinion on crap before hassling us about it. Once you have passed the moderation period (think of it as a test), you will be able to post normally, just like all the other retards.

A question to indie RPG developers

AvengerDr

Novice
Joined
Jun 23, 2007
Messages
15
During the development of your game I assume that at a certain point you had to implement methods and algorithms to automate random NPC stat generation, combat computations and other "common" tasks. You had to develop your own rule system for that purpose. I was wondering: if some sort of "AD&D++" or "GURPS#" library was available that you could use in your code, would you have used that instead of developing your onw system?

I was asking myself that because for my future 4x space strategy game that I'm developing, sooner or later I'll have to develop some sort of RPG-like rule system. For example to handle combat between spaceships I could either use a heartless approach and use rays, intersections, inertial momentum, collisions and compute combat results in that way or use a more "humane" approach and leave it to the pilot's "dogfighting" skill: nothing that a dice roll can't solve. I also like to have each spaceship have a randomly generated NPC pilot, whose stat will be used in combat. Pilots who gain so much experience behind the scenes could be "promoted" to "special" NPCs as those in Moo2 or Rome Total War, that you could then use as governors.

Anyway back to my original question. To handle that, I'll have to develop my own rule system, and although I'd have used a "GURPS#" library if it were available freely, I'll have to develop my own system. So I was interested to know from you "indie" developers if, some sort of such library (that offered classes to handle NPC generation, combat resolution and so on), publicly available, existed, would you consider using it?

The library as I envision it would have multiple modules, like sourcebooks. There could be specialized DLLs to handle space combat, others to handle modern ranged combat, fantasy ones and so on.

One point to be made is that obviously every game handles combat in its very own way. In fact I don't mean for the library to have algorithms to resolve combat on its own, just algorithms to check whether a character can "hit" another depending on its stat. The equivalent of a THAC0 computation for example. It would be you who decide where and when to use that function call.

I'd obviously be tempted to answer "No, I'd prefer to make my own rule system", but that may be time consuming, and as I said, if a GURPS# (I'm a GURPS and C# enthusiast as you have imagined :D) were available I'd use that, both for "brand recognition" and for the fact that I could be able to concentrate more on the game itself without having to spend time to balance a rule system that I can already assume to be as good as it can get.
I'm sure that a community effort could design a good computer-usable rule system and who knows, maybe even gain sufficient recognition to be used in some new indie projects.

What is your opinion on this idea?
 

Lemunde

Scholar
Joined
Jan 16, 2006
Messages
322
For me personally, no. It's much easier to create my own system that I know will work in the engine I've designed than to try to adapt a premade system that I'm not 100 percent familiar with. It's hard to tell you why you shouldn't use a system like that in your own game other than it would take all the fun out of making the game.

If you use your own system then you can call the game yours and not just an adaptation of a table top game. You can also freely implement your own features into the system.

My advice on your 4x game is to try and avoid overly complex systems that are transparent to the player. Most players aren't physics majors and would rather not strain their brains on something that is only a small part of the game. That's not to say the system shouldn't be complex at all but you should stick with definitions that most players can understand.

The one turn based strategy game that I thought had the best combat was Age of Wonders. Instead of overloading each unit with stats each unit only had four stats and several abilities. The stats were simply health, attack rating, defense rating and movement speed. The abilities ranged from spell casting to mountain climbing to first strike.

This kind of made the combat feel like a sort of Magic the Gathing game and it worked really well. It was complex enough to provide a wide variety of strategies but simple enough to understand at a glance what each unit was capable of.
 

Jasede

Arcane
Patron
Joined
Jan 4, 2005
Messages
24,793
Insert Title Here RPG Wokedex Codex Year of the Donut I'm very into cock and ball torture
It's just a cheap knockoff of Master of Magic, which had even better combat. AoW was the only game that "felt" like MoM... not exactly, but damn close.

I'd kill people for MoM 2. Well. Only criminals, of course.
 

Psilon

Erudite
Joined
Feb 15, 2003
Messages
2,018
Location
Codex retirement
I've discussed this before with others, and the gist is it's probably not worth it. Especially for a 4X game, where individual NPCs are generally meaningless compared to ship combat (which had better be designed for the specific game) and mass troop battles (which are of necessity abstracted all to hell). "Generic" rule systems tend to be utter nightmares to integrate into a preexisting code base. Here are some of the problems we came up with:

Multiple sourcebook modules basically equate to a zillion plugins that all have to be initialized in the correct order, and this will probably result in tons and tons of boilerplate explosion-at-the-pattern-factory code. Especially when you consider that no two developers are likely to have the same engine architecture, this will mean lots of translation code as you marshal from GameA representations to GURPS# representations and vice versa.

Brand recognition is unfortunately a non-issue with computer RPG systems. The only rulesets that are worth promoting are basically impossible to get the licenses for. You use GURPS as an example, but the only GURPS implementation ever attempted (Fallout) got the license pulled for violence reasons. Wizards has caught flak for the bad-guys-always-lose policy they attach to D&D, but it hasn't changed yet. White Wolf, well, not only are they 0-for-2 on high-selling Storyteller games, but they've got the only-for-Goths stigma attached to their system.

The computer game rulesets have zero name recognition. Since historically "reusing a ruleset" means "reusing an engine," there isn't the long tradition of "Uses X Rules!" for non-PnP systems. In fact, it usually causes derision for "recycling the same old crap." (See: Hammer & Sickle, Gold Box games, any Spiderweb game after Avernum 1.) Even the Wizardry series kept revising and redoing along the way.

Plus, keep in mind that all of these sourcebooks would have to be really, really generic. As in, "so generic you're rewriting most of the rules anyway." GURPS shoots for four stats, ST, DX, IQ, and HT. How many of those are actually useful in a 4X? Once you put someone's ass in a pilot's seat there is absolutely no point to ST or HT until the ship is hammered to pieces by incoming fire. Do you use a grid or continuous motion? What type of grid?

Do you use an advantage/flaw system? If so, how can you meaningfully generate random NPCs using an algorithm that has no idea whether certain combinations make sense? I don't mean Perfect Eyesight and Blindness, but I could see a random generator not really knowing the difference between Appearance: Supermodel, Eidetic Memory, and Cast Iron Gut.

Factor in the lack of a human overseer, and I'd say skip it. A freely available rule set is one thing, and might be fun, but it should be more akin to the D20 SRD than to GTK+. There are just too many things that have to be done to fit a system to a game.
 

Vault Dweller

Commissar, Red Star Studio
Developer
Joined
Jan 7, 2003
Messages
28,024
AvengerDr said:
I was wondering: if some sort of "AD&D++" or "GURPS#" library was available that you could use in your code, would you have used that instead of developing your onw system?
No. I'm in favor of creativity and originality, so I strongly dislike all and any "borrowed" elements.
 

Sovy Kurosei

Erudite
Joined
Dec 29, 2004
Messages
1,535
AvengerDr said:
I'd obviously be tempted to answer "No, I'd prefer to make my own rule system", but that may be time consuming, and as I said, if a GURPS# (I'm a GURPS and C# enthusiast as you have imagined ) were available I'd use that, both for "brand recognition" and for the fact that I could be able to concentrate more on the game itself without having to spend time to balance a rule system that I can already assume to be as good as it can get.

Unless you were intimately familiar with that ruleset in the first place I don't think you'll save as much time. Making a new ruleset is no more complicating than elementary math, some spreadsheets, graphs and a bit of white box testing. You can design and change your custom made ruleset to fit your game instead of trying to forcing an existing ruleset in.
 

AvengerDr

Novice
Joined
Jun 23, 2007
Messages
15
Yours are all valid points. I plan to have some RPG elements in the game, if only to avoid having over-complicated math to handle combat. If I can implement my ideas about combat, there will be minimal player involvement.

Yes, it's true that a ST of a pilot does not matter while in a cockpit, but it's piloting or gunnery skill should for example. True, the player wouldn't be able to see the character sheet of every pilot in the galaxy but they would be internally needed to manage combat. Otherwise the alternative is to do it "the hard way" with mathematical computations still hidden to the player (otherwise the game would become an advanced math tutorial). Also I'd like for some of these "pilots" to "ascension" to "reality" by becoming "real" npcs, as it happened in Rome Total War, so that you could use them as governors or in covert missions. I've seen many times people say "if it's hidden from the player, why have it in the first place?". If game mechanics are hidden, then how is the game going to work? :D The AI can't yet tell you a story in front of a campfire, so to speak Obviously this is an extremization but...

As for the random NPC generator, well obviously it would not be totally random. Otherwise a spaceship pilot could be a master iin the use of plasma rifles and also very skilled in the use of crossbows.. That obviously doesn't make sense. The random generator would generate random NPC pilots within the limits defined by some sort of sci-fi pilot template.

It is not my intent to develop a "generic ruleset" because that would draw away time from my original intent, which is to develop a 4x game. But since I'll have to develop some sort of rules to handle combat, special character stats and skills I was wondering if it would be a good idea to make that library freely available. Odds are that no one would use it, but hey you never know :)
 

Psilon

Erudite
Joined
Feb 15, 2003
Messages
2,018
Location
Codex retirement
I question whether you really need an "RPG" rule set for combat in the first place. This is a 4X, yes? Akin to GalCiv or Master of Orion? If so, ship combat should be stolen from wargames, not RPGs! RPG vehicle combat is for sissies who can't handle switching to Car Wars or Star Fleet Battles. Seriously, just give each component/hull an attack/defense strength and do things that way. Ship design is always more fun than colony management.

Assume that, barring the occasional Zap Brannigan, the various planetary military academies will produce pilots, gunners, and tacticians who are all reasonably competent plus or minus some minor random correction factor for morale and training. If you want Glorious Leaders, just assign an XP counter to each vessel. If it passes some threshold, start doing the random checks for leader promotion. Then and only then should you bother with NPC stats. Until then, it's not worth it.

As for "character sheets of every pilot in the galaxy being internally needed to manage combat," how is THAT not the hard way? Seriously, by the time you've figured out how all the different crew members interact... you still have one set of stats for a single fighter squadron. It's just derived in a far more complicated fashion than Attack/Defense/Firepower/HP.
 

AvengerDr

Novice
Joined
Jun 23, 2007
Messages
15
In fact that's how I'm intending to proceed. Having "pilots" not become "real NPCs" until some random event happens. For the particular case of my game I intend to have spaceships have their own stats (which will be the result of your choices in the canonical ship design screen). These (spaceship) stats will be the primary actors in deciding the outcome of combat. I just wanted to modulate it by not-so random chance.

What I mean is that if you only leave the combat to the stats of the spaceships, the outcome will be very deterministic in nature. In moo2 there was the extreme that if you had the "initiative" and your ships could destroy the enemy ones in a single shot, combat could end in only one turn (provided that you had the required number of ships).

So my idea is to have "second class npcs" (pilots, infantry battalllions, and so on) have just a couple of stats relevant to their job. For a pilot they could be, piloting, gunnery, bravery and morale for example. By pilot I mean the abstraction of the crew of the unit. If a ship it is said to require more than one people inside, it's not like I'm going to model the copilot and the mechanic and the hidden intruder aboard.

In this way a faster/better/ship will often have an advante over an older model one, but the pilot's bravery could prove to be a deciding factor. It doesn't seem to be so hard to me... just a couple more numbers that could help in providing more variance. What if the last crat defending of the capital ship is a coward pilot :D?

Eventually if pilots/admirals make the jump to "real npcs" that you can use as governors/spies/hero character, they'd gain a more complete set of stats.
 

Joe Krow

Erudite
Joined
Feb 16, 2007
Messages
1,162
Location
Den of stinking evil.
Jasede said:
It's just a cheap knockoff of Master of Magic, which had even better combat. AoW was the only game that "felt" like MoM... not exactly, but damn close.

I'd kill people for MoM 2. Well. Only criminals, of course.

Agreed. If someone handed me a box with a button on it and said "if you press this button I will give you MoM 2 but someone you don't know will die" I would definitely press the button.
 

jcompton

Novice
Joined
Apr 8, 2004
Messages
45
I doubt this proposal would be terribly valuable to many developers. After all, someone looking just for "good examples of procedural creature creation" need search no further than their local open-source roguelike--those guys have been at it for decades.

Plus which, by time people translated the stats used in your system to the ones their system required, they would probably simply be better off developing their own algorithms to suit the strengths and intricacies of their system.

Bottom line--although good procedural creature generation is potentially a good selling point of a complete engine, it has limited applicability in a more generic setting, unless your ruleset is so appealing and useful that they are willing to take the ruleset and the random generator together.
 

As an Amazon Associate, rpgcodex.net earns from qualifying purchases.
Back
Top Bottom